updates

utils/gram2vec_feat_utils.py

@@ -144,7 +144,7 @@ def show_combined_spans_all(selected_feature_llm, selected_feature_g2v,
     background_authors_embeddings_df = background_authors_embeddings_df[background_authors_embeddings_df.authorID.isin(visible_authors)]
     background_and_task_authors = pd.concat([task_authors_embeddings_df, background_authors_embeddings_df])
 
-    authors_texts = ['\n\n =========== \n\n'.join(x) if type(x) == list else x for x in background_and_task_authors[:max_num_authors]['fullText'].tolist()]
+    authors_texts = ['\n\n'.join(x) if type(x) == list else x for x in background_and_task_authors[:max_num_authors]['fullText'].tolist()]
     authors_names = background_and_task_authors[:max_num_authors]['authorID'].tolist()
     print(f"Number of authors to show: {len(authors_texts)}")
     print(f"Authors names: {authors_names}")

utils/ui.py

@@ -162,8 +162,8 @@ def task_HTML(mystery_text, candidate_texts, predicted_author, ground_truth_auth
 
     # if any of the texts is a list of text then concatenate them
     if isinstance(mystery_text, list):
-        mystery_text = "\n\n".join(["Text: {}".format(x) for x in mystery_text])
-        candidate_texts = ["\n\n".join(["Text: {}".format(t) for t in x]) for x in candidate_texts]
+        mystery_text = "\n\n".join(mystery_text)
+        candidate_texts = ["\n\n".join(x) for x in candidate_texts]
 
     header_html = f"""
     <div style="border:1px solid #ccc; padding:10px; margin-bottom:10px;">

utils/visualizations.py

@@ -305,7 +305,7 @@ def handle_zoom(event_json, bg_proj, bg_lbls, clustered_authors_df, task_authors
         task_only_df = task_authors_df
 
     def _to_text(x):
-        return '\n\n =========== \n\n'.join(x) if isinstance(x, list) else x
+        return '\n\n'.join(x) if isinstance(x, list) else x
 
     task_texts = [_to_text(x) for x in task_only_df['fullText'].tolist()]